基于信息熵和模态特性的在役混凝土桥梁裂缝扩展机理与诊断方法研究

With demand of economic development, increasing traffic and overloaded transportation trend, the existing concrete bridge as a control and key part of traffic system is more and more threatened by crack damage. This project is proposed to study crack developing mechanism and diagnosing method of existing concrete bridge，by combining modal characteristics with information entropy theory of dissipative systems. Firstly, based on extended finite element method, the energy distribution and evolutive rule is analyzed at the elements of concrete crack expansion region, to propose the crack evolution tracing factor based on information entropy of dissipation system, which can track crack propagation process by establishing crack evolution entropy function, and popularizing the result into use of existing concrete bridge structure under multi-factor action. The weight coefficient of various influencing factors will be obtained by regression analysis; the crack extension mechanism will be studied under the action of single factor and multi-factor coupling, to expound the law of crack entropy change. Then, by introducing modal characteristics, proposing modal strain energy entropy difference and modal flexibility entropy as indicates to establish diagnosis method, which two indexes can improve the diagnostic accuracy and anti-noise performance in the aspect of crack meso-scale, and are analyzed and validated by coupling with result of numerical example, model test and bridge field test. Lastly, through the above researches, to reveal the internal rules of crack developing, the reasonable and effective crack diagnosis method will be put forward accordingly, to provide theoretical support and scientific basis in disease warning and control for existing concrete bridge.

随着经济发展的需求，交通量日益增长，运输工具呈现重载化趋势，在役混凝土桥梁作为交通系统的关键和控制部位，越来越多的受到裂缝损伤的威胁。本项目拟结合耗散结构信息熵理论与模态特性的各自优点，对在役混凝土桥梁裂缝的扩展机理和诊断方法开展研究。首先采用扩展有限元方法，分析混凝土裂缝扩展单元的能量分布，提出基于耗散结构信息熵的裂缝演化示踪因子，推导裂缝演化熵函数，追踪裂缝扩展过程，并推及到多因素作用下在役混凝土桥梁结构，回归分析各影响因素的权重系数，研究单因素和多因素耦合作用下裂缝扩展机理，阐明熵变规律；进一步引入模态特性，提出并建立以模态应变能熵值差与模态柔度熵为指标的诊断方法，以提高裂缝细观尺度的诊断精度与抗噪性能，再结合数值算例、模型和实桥试验结果进行对比分析和验证。最后通过以上研究，揭示裂缝扩展内在规律，建立合理有效的裂缝诊断方法，为在役混凝土桥梁病害预警与防治提供理论支撑和科学依据

本项目拟定的目标已经基本完成，即通过信息熵理论和结构动力特性的混凝土桥梁裂缝（损伤）识别框架，最终实现损伤定位和损伤定量。具体工作如下：将裂缝开展的最终形态视为结构损伤，引入模态特性，建立了不同的能量示踪因子，包括模态曲率信息熵、柔度曲率信息熵等。解释了部分示踪因子的物理意义，并结合小波包、神经网络、信息融合等技术实现桥梁结构静、动态模态信息相结合，搭建了不同示踪因子下几种完整、准确的损伤诊断方法。通过有限元分析、缩尺模型试验和实桥试验对发展的损伤诊断框架进行考察。各项子研究结果表明，结构动力指纹与结构信息熵的结合是一种新颖且有潜力的损伤识别技术。在此基础上所发展的损伤诊断框架能够准确的对混凝土桥梁结构进行损伤定位和损伤定量；为结构损伤识别提供了新的视角；为在役混凝土桥梁病害预警与防治理论支撑和科学依据。